Tag: 系统架构 (7 articles)

Through Lalit Maganti's experience, it reveals the potential and limitations of AI in software development, particularly the challenges in architectural design.

Continual learning for AI agents is not just about updating model weights; crucial evolution happens at the 'harness' and 'context' layers, offering new ways to build truly personalized and growing agents.

DeepSeek V4 achieves efficient million-token long-context inference on vLLM through innovative KV cache compression and sparse attention mechanisms, marking a new era for long-text processing.

vLLM Semantic Router discovered that its vision encoder signals were significantly misaligned with the reference model, causing confidently wrong routing decisions, which reveals that signal correctness becomes a critical control-plane requirement as AI systems evolve from processing text to full requests.

Meta is strengthening end-to-end encrypted backups for WhatsApp and Messenger using hardware security modules (HSMs) and transparent deployment, making backups inaccessible even to Meta itself, with independent third-party audits.

A developer runs the 'For You' recommendation algorithm for 72,000 users on a living room gaming PC for $30 a month, revealing new possibilities for algorithmic democratization.

vLLM and Novita AI introduce PegaFlow, an external KV cache service that decouples cache from the inference process, dramatically improving startup speed, throughput, and resource efficiency for production LLM serving.